TomatoDIFF: On-plant Tomato Segmentation with Denoising Diffusion Models

Abstract

Artificial intelligence applications enable farmers to optimize crop growth and production while reducing costs and environmental impact. Computer vision-based algorithms in particular, are commonly used for fruit segmentation, enabling in-depth analysis of the harvest quality and accurate yield estimation. In this paper, we propose TomatoDIFF, a novel diffusion-based model for semantic segmentation of on-plant tomatoes. When evaluated against other competitive methods, our model demonstrates state-of-the-art (SOTA) performance, even in challenging environments with highly occluded fruits. Additionally, we introduce Tomatopia, a new, large and challenging dataset of greenhouse tomatoes. The dataset comprises high-resolution RGB-D images and pixel-level annotations of the fruits.

Plant phenotyping relevance

植物上のトマト果実を画像からセグメンテーションする手法を開発・比較し、RGB-D画像と画素アノテーションのデータセットも提供しており、植物器官の状態・位置推定に関わる方法が中心である。

Code and data availability

The paper introduces TomatoDIFF and the Tomatopia dataset, with explicit public availability of source code and dataset at the authors' GitHub repository. It also trains/evaluates on the public Kaggle 'Tomato dataset' (andrewmvd/tomato-detection), which is a paper-specific public image dataset used directly in the phen